To The Editor:
After reading the article by Colwell et al., "Oral Direct Thrombin
Inhibitor Ximelagatran Compared with Warfarin for the Prevention of Venous
Thromboembolism After Total Knee Arthroplasty"
(2005;87:2169-77), we are
concerned that the facts presented by the authors do not appear to justify
their conclusions and that some important parameters of patient evaluations
are lacking.
In this report, the authors note that 0.3% of the patients in the
ximelagatran group died from a cause in which pulmonary embolism could not be
ruled out whereas there were no deaths from pulmonary embolism in the warfarin
group. This incidence is relatively high when compared with that reported in
most recent studies, which have demonstrated a 0.1% incidence of fatal
pulmonary embolism following total joint surgery. More detailed information
regarding these deaths would be helpful to assist the reader in
differentiating the patients in whom pulmonary embolism could not be ruled out
from patients with fatal pulmonary embolism.
Major bleeding complications were noted in 1% of the patients in the
ximelagatran group, compared with 0.4% of those in the warfarin group. Major
bleeding was defined as critical site bleeding, such as intracranial or
retroperitoneal bleeding; bleeding requiring surgical intervention; or fatal
bleeding. In the ximelagatran group, major bleeding episodes included one
fatal gastrointestinal bleed and one subdural hematoma. All other recorded
bleeding episodes were considered minor. Although the number of patients with
major bleeding episodes was small, it was more than twice that in the warfarin
group. If this study had been fivefold larger, the difference could have been
significant. With more than 300,000 total knee replacements being done each
year, these small differences can become clinically and statistically
important.
In addition, and most importantly, there is no mention of the clinical
outcomes or how the major and/or minor bleeds affected patient health.
The venographic data in the article showed no difference between the two
drugs in terms of the incidence of proximal deep venous thromboses, the more
worrisome clots. They also showed only a 9.7% diminution in the incidence of
distal deep venous thrombosis in the ximelagatran group, a finding that in our
opinion is of dubious importance.
Therefore, in this report, ximelagatran was associated with a 0.3% rate of
fatal pulmonary embolism whereas warfarin was associated with a 0% rate;
ximelagatran was associated with a bleeding rate that was more than two times
higher than that associated with warfarin; and the two drugs were associated
with an equal incidence of proximal deep venous thrombosis. The only finding
in favor of ximelagatran was a limited reduction in distal deep venous
thrombosis, which may be of little clinical importance.
Yet, in the abstract, the authors conclude that the new drug
"demonstrates superior efficacy compared with warfarin prophylaxis, with
no wound complications and no significant difference with respect to bleeding
events." We are not persuaded that the new drug warrants this
conclusion. To the contrary, ximelagatran appears to be no more effective than
warfarin, and, if a greater number of patients had been included in the study,
a significantly higher incidence of patient-related complications and adverse
events might have been revealed.
For the above reasons, we remain concerned about the conclusions published
in this article.